CN109167998A - Detect method and device, the electronic equipment, storage medium of camera status - Google Patents
Detect method and device, the electronic equipment, storage medium of camera status Download PDFInfo
- Publication number
- CN109167998A CN109167998A CN201811377288.9A CN201811377288A CN109167998A CN 109167998 A CN109167998 A CN 109167998A CN 201811377288 A CN201811377288 A CN 201811377288A CN 109167998 A CN109167998 A CN 109167998A
- Authority
- CN
- China
- Prior art keywords
- camera
- picture
- type
- convolutional neural
- neural networks
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N17/00—Diagnosis, testing or measuring for television systems or their details
- H04N17/002—Diagnosis, testing or measuring for television systems or their details for television cameras
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- General Health & Medical Sciences (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Image Analysis (AREA)
- Studio Devices (AREA)
Abstract
The present embodiments relate to image identification technical fields, disclose method and device, the electronic equipment, storage medium of a kind of detection camera status.In the present invention, by the picture for obtaining the N seed type that camera takes, wherein N is the natural number greater than 2；Convolutional neural networks model training is carried out using the picture of N seed type as training data；Utilize the type of the picture of trained model prediction camera captured in real-time；Determine whether camera can be used according to prediction result；A kind of method of reliable detection camera status is provided, and is classified with the picture that convolutional neural networks shoot camera, may make judgement whether available to camera more accurate.
Description
Technical field
The present embodiments relate to image identification technical field, in particular to a kind of method for detecting camera status.
Background technique
In recent years, self-service machine is a dark horse, the deep welcome by user.Newest self-service machine technology abandons gravity
Induction, using Machine Vision Recognition commodity, so that in the case where unattended, user can be required for actual touch chooses oneself
Commodity, and can be settled accounts automatically after user's shopping, realize that taking is to walk, eliminate user's cash or mobile phone and carry out
The process of payment.
At least there are the following problems in the prior art for inventor's discovery:
When the camera lens of self-service machine haze, self-service system can not be known in time, to can not accurately know
Other commodity.
Summary of the invention
A kind of method and device for being designed to provide detection camera status of embodiment of the present invention, is deposited electronic equipment
Storage media enables the system to the state for accurately detecting camera.
In order to solve the above technical problems, embodiments of the present invention provide a kind of method for detecting camera status, including
Following steps: the picture for the N seed type that camera takes is obtained, wherein N is the natural number greater than 2；By the picture of N seed type
Convolutional neural networks model training is carried out as training data；Utilize the picture of trained model prediction camera captured in real-time
Type；Determine whether camera can be used according to prediction result.
Embodiments of the present invention additionally provide a kind of device for detecting camera status, comprising: module are obtained, for obtaining
The picture for the N seed type that camera takes, wherein N is the natural number greater than 2；Training module, for by the picture of N seed type
Convolutional neural networks model training is carried out as training data；Prediction module, for real using trained model prediction camera
When the type of picture that shoots；Determination module, for determining whether camera can be used according to prediction result.
Embodiments of the present invention additionally provide a kind of electronic equipment, comprising: at least one processor；And at least
The memory of one processor communication connection；Wherein, memory is stored with the instruction that can be executed by least one processor, instruction
It is executed by least one processor, so that at least one processor is able to carry out the method such as above-mentioned detection camera status.
Embodiments of the present invention additionally provide a kind of computer readable storage medium, are stored with computer program, calculate
Machine program realizes above-mentioned detection camera status method when being executed by processor.
A plurality of types of pictures that embodiment of the present invention takes in terms of existing technologies, by obtaining camera,
And these pictures is used to carry out convolutional neural networks model training as training data, reliable disaggregated model can be obtained, is utilized
The type of the picture of trained model prediction camera captured in real-time can be predicted the type of the photo of camera shooting, according to pre-
It surveys as a result, can determine whether camera can be used, also, convolutional neural networks model can accurately determine the type of each picture,
And the type of picture is more, the judgement of camera available mode is also just more accurate.Present embodiments provide a kind of reliable detection
The method of camera status, and classifying with the picture that convolutional neural networks shoot camera, may make to camera whether
Available judgement is more accurate.
In addition, above-mentioned carry out convolutional neural networks model training for the picture of N seed type as training data, it is specific to wrap
It includes: selecting multiple convolutional neural networks；Using the picture of N seed type as training data to each volume in multiple convolutional neural networks
Product neural network carries out model training；The type of the above-mentioned picture using trained model prediction camera captured in real-time, specifically
It include: the type using the picture of trained each convolutional neural networks model prediction camera captured in real-time；According to each convolution mind
Prediction result through network model is voted, and final prediction result is obtained.Present embodiments provide for a kind of specific fortune
The method that the type of the picture of camera captured in real-time is predicted with convolutional neural networks, by being carried out to multiple convolutional neural networks
Multiple prediction models can be obtained in model training, obtain final prediction by being voted according to the prediction result of multiple models
As a result, prediction result deviation is larger caused by can avoid the exception because of some model, so that prediction result is more accurate reliable.
In addition, above-mentioned multiple convolutional neural networks include at least depth residual error network；It is above-mentioned according to each convolutional neural networks
Prediction result vote, obtain final prediction result, comprising: if voting results be flat ticket, with depth residual error network
Prediction result as final prediction result.A kind of solution when voting results are flat ticket is present embodiments provided,
Since the prediction result accuracy rate of depth residual error network is higher, when there is flat ticket, made with the prediction result of depth residual error network
It can guarantee to more maximum probability the accuracy of prediction result for final prediction result.
In addition, determining whether camera can be used according to prediction result specifically: above-mentioned whether to determine camera according to prediction result
Can use specifically: judge prediction result whether characterize camera captured in real-time picture it is clear, it is no if so, determine that camera is available
Then, determine that camera is unavailable.A kind of specific judgement whether available method of camera is present embodiments provided, due to actually answering
With in the process, the service quality of self-service system will be influenced whether when the blurring of photos of camera shooting, is shot by camera
Picture whether clearly determine whether camera can be used, be a kind of realistic criterion.
In addition, the type of the above-mentioned picture using trained model prediction camera captured in real-time, specifically includes: utilizing instruction
The type for the L picture that the model prediction camera perfected is shot recently, wherein L is the natural number greater than 1；Above-mentioned judgement prediction
As a result the picture for whether characterizing camera captured in real-time is clear, if so, it is available to determine camera, otherwise, it is determined that camera is unavailable, and tool
Body are as follows: it is clear to judge whether prediction result characterizes L picture, if so, determining that camera is available, otherwise, it is determined that camera can not
With.In the present embodiment, when determining whether camera available, the plurality of pictures that camera is shot recently is predicted, only when
When plurality of pictures is clear, just determine that camera is available, so that the judgement to camera status is more accurate.
In addition, then the down state of the camera is reported after the above-mentioned judgement camera is unavailable.The present embodiment
In, the down state of camera is reported, related personnel can be notified to handle camera in time.
In addition, above-mentioned N is 6, and above-mentioned N seed type, specifically: empty clear type, empty vague category identifier, clearest class
Type, most vague category identifier expose clear type, expose vague category identifier.Present embodiments provide a kind of tool of the picture of camera shooting
Body type, this six seed type is camera type common during shooting photo, also, convolutional neural networks can be well
This six seed type is distinguished, so that the detection of camera status is more accurate.
Detailed description of the invention
One or more embodiments are illustrated by the picture in corresponding attached drawing, these exemplary theorys
The bright restriction not constituted to embodiment.
Fig. 1 is the method flow diagram for the detection camera status that first embodiment provides according to the present invention；
Fig. 2 is the method flow diagram for the detection camera status that second embodiment provides according to the present invention；
Fig. 3 is the method flow diagram for the detection camera status that third embodiment provides according to the present invention；
Fig. 4 is the apparatus structure schematic diagram for the detection camera status that the 4th embodiment provides according to the present invention；
Fig. 5 is the electronic devices structure schematic diagram that the 5th embodiment provides according to the present invention.
Specific embodiment
In order to make the object, technical scheme and advantages of the embodiment of the invention clearer, below in conjunction with attached drawing to the present invention
Each embodiment be explained in detail.However, it will be understood by those skilled in the art that in each embodiment party of the present invention
In formula, in order to make the reader understand this application better, many technical details are proposed.But even if without these technical details
And various changes and modifications based on the following respective embodiments, the application technical solution claimed also may be implemented.With
Under the division of each embodiment be for convenience, any restriction should not to be constituted to specific implementation of the invention, it is each
Embodiment can be combined with each other mutual reference under the premise of reconcilable.
The first embodiment of the present invention is related to a kind of methods for detecting camera status.In present embodiment, camera is obtained
The picture of the N seed type taken, wherein N is the natural number greater than 2；The picture of N seed type is rolled up as training data
Product neural network model training；Utilize the type of the picture of trained model prediction camera captured in real-time；According to prediction result
Determine whether camera can be used, a plurality of types of pictures taken by obtaining camera, and use these pictures as training data
Convolutional neural networks model training is carried out, reliable disaggregated model can be obtained, it is real-time using trained model prediction camera
Whether the type of the type of the picture of shooting, the photo of the predictable shooting of camera out can determine camera according to prediction result
It can use, also, convolutional neural networks model can accurately determine the type of each picture, and the type of picture is more, camera is available
The judgement of state is also just more accurate.The method of detection camera status in present embodiment is as shown in Figure 1, below to this implementation
The realization details of the method for the detection camera status of mode is specifically described, and the following contents is only for convenience of the reality for understanding offer
Existing details, not implements the necessary of this programme.
Step 101: obtaining the picture for the N seed type that camera takes.
Specifically, computer can obtain a large amount of pictures that camera is shot from the memory of camera or cloud, and pass through people
These pictures are divided into N class by the mode of work, wherein N is the natural number greater than 2.
In a specific self-service airport scape, above-mentioned N can be specially 6, above-mentioned N seed type, specifically: it is empty clear
Clear type, empty vague category identifier, clearest type, most vague category identifier expose clear type, expose vague category identifier.This six type
Type is camera type common during shooting photo, carries out analyzing the data that can guarantee acquisition to the picture of these types
Multiplicity enough.
Step 102: carrying out convolutional neural networks model training for the picture of N seed type as training data.
Specifically, computer can extract two-dimensional array (gray level image) or three-dimensional array from every picture of acquisition
(color image), and using the array of extraction as the input of convolutional neural networks, using the type of the picture as corresponding output
Carry out model training (the differentiation feature of the extractable image out of convolutional neural networks is so that classifier is learnt).
Step 103: utilizing the type of the picture of trained model prediction camera captured in real-time.
Specifically, computer obtains the picture of camera captured in real-time, and corresponding characteristic is extracted, utilizes training
Good convolutional neural networks model predicts that it is any in N seed type that the picture particularly belongs to.
Step 104: determining whether camera can be used according to prediction result.
Specifically, computer can predict the picture of camera captured in real-time using trained convolutional neural networks model
Type determine that camera is unavailable if prediction result shows that picture belongs to Exception Type (such as picture very fuzzy).
Present embodiment compared with the prior art for, by obtaining a plurality of types of pictures for taking of camera, and use this
A little pictures can obtain reliable disaggregated model as training data progress convolutional neural networks model training, using training
Model prediction camera captured in real-time picture type, can be predicted the type of the photo of camera shooting, according to prediction result,
It can determine whether camera can be used, also, convolutional neural networks model can accurately determine the type of each picture, and picture
Type is more, and the judgement of camera available mode is also just more accurate.Present embodiments provide a kind of reliable detection camera status
Method, and classify with the picture that convolutional neural networks shoot camera, may make and whether available to camera sentence
It is fixed more accurate.
Second embodiment of the present invention is related to a kind of method for detecting camera status.Second embodiment is implemented with first
Mode is roughly the same, is in place of the main distinction: in second embodiment of the invention, providing pre- with convolutional neural networks
The method for surveying the type of the picture of camera captured in real-time.The flow chart of the present embodiment is as shown in Fig. 2, specifically described below.
Step 201: obtaining the picture for the N seed type that camera takes.
Step 201 is roughly the same with the step 101 of first embodiment, to avoid repeating, no longer repeats one by one here.
Step 202: selecting multiple convolutional neural networks.
Specifically, common convolutional neural networks are there are many network structure, for example, VGG16, GoogLeNet and
ResNet, mobilenet etc. therefrom select the convolutional neural networks of multiple and different network architectures.
Step 203: using the picture of N seed type as training data to each convolutional Neural net in multiple convolutional neural networks
Network carries out model training.
Specifically, computer is using the data extracted from every picture as the input of each convolutional neural networks,
Each convolutional neural networks model training is carried out using the type of the picture as corresponding output.
Step 204: utilizing the type of the picture of trained each convolutional neural networks model prediction camera captured in real-time.
Specifically, computer is using each model in trained multiple convolutional neural networks models, it is real to camera
When the type of picture that shoots predicted.
Step 205: being voted according to the prediction result of each convolutional neural networks model, obtain final prediction result.
Specifically, the prediction result due to each convolutional neural networks model may be different, at this moment, according to the pre- of each model
Result is surveyed to vote, for example, a total of 5 convolutional neural networks models, wherein the prediction knot of 3 convolutional neural networks models
Fruit be A class, in addition the prediction result of 2 convolutional neural networks models be B class, then determine final prediction result for A class, i.e., most
Determine that the type of the picture of camera captured in real-time is A class eventually.
It is noted that multiple convolutional neural networks include at least depth residual error network (i.e. ResNet)；According to each volume
The prediction result of product neural network is voted, and final prediction result is obtained, comprising: if voting results are flat ticket, with depth
The prediction result of residual error network is spent as final prediction result.Since the prediction result accuracy rate of depth residual error network is higher,
When there is flat ticket, it can guarantee to more maximum probability prediction knot using the prediction result of depth residual error network as final prediction result
The accuracy of fruit.
Step 206: determining whether camera can be used according to prediction result.
Step 206 is roughly the same with the step 104 in first embodiment, to avoid repeating, no longer repeats one by one here.
Present embodiment compared with the prior art for, provide a kind of specifically real with convolutional neural networks prediction camera
When the method for the type of picture that shoots multiple predictions can be obtained by carrying out model training to multiple convolutional neural networks
Model obtains final prediction result by being voted according to the prediction result of multiple models, can avoid because of some model
Prediction result deviation is larger caused by exception, so that prediction result is more accurate reliable.
Third embodiment of the present invention is related to a kind of method for detecting camera status.Third embodiment is implemented with first
Mode is roughly the same, is in place of the main distinction: in third embodiment of the invention, providing a kind of specific judgement camera
Whether available method.The flow chart of the present embodiment is as shown in figure 3, specifically described below.
Step 301: obtaining the picture for the N seed type that camera takes.
Step 302: carrying out convolutional neural networks model training for the picture of N seed type as training data.
Step 303: utilizing the type of the picture of trained model prediction camera captured in real-time.
Step 301 is roughly the same to step 103 with the step 101 of first embodiment to step 303, to avoid repeating,
Here it no longer repeats one by one.
Step 304: judge prediction result whether characterize camera captured in real-time picture it is clear, if so, thening follow the steps
305, it is no to then follow the steps 306.
Specifically, the type of the photo of camera shooting is divided into empty clear type, empty in the scape of self-service airport
Vague category identifier, clearest type, most vague category identifier expose clear type, six kinds of vague category identifier are exposed, when convolutional neural networks
When prediction result shows that the type of photo is one such for empty clear type, clearest type, the clear type of exposure, determine
Camera is available, otherwise determines that camera is unavailable.
It is noted that the type of the above-mentioned picture using trained model prediction camera captured in real-time, specific to wrap
It includes: utilizing the type for the L picture that trained model prediction camera is shot recently, wherein L is the natural number greater than 1；It is above-mentioned
Judge prediction result whether characterize camera captured in real-time picture it is clear, if so, determining that camera is available, otherwise, it is determined that camera
It is unavailable, specifically: it is clear to judge whether prediction result characterizes L picture, if so, determine that camera is available, otherwise, it is determined that
Camera is unavailable.In the present embodiment, when determining whether camera is available, the plurality of pictures that camera is shot recently is carried out pre-
It surveys, only when plurality of pictures is clear, just determines that camera is available, so that the judgement to camera status is more accurate.
Step 305: determining that camera is available.
Step 306: determining that camera is unavailable.
Specifically, then show that camera lens have hazed when the picture blur of prediction result characterization camera captured in real-time,
Unmanned selling system has the risk that can not accurately identify commodity, determines that camera is unavailable at this time, and by the down state of camera
It reports, relevant staff is enabled to handle the situation as early as possible.
Present embodiment compared with the prior art for, provide it is a kind of specific determine the whether available method of camera, by
In in actual application, the service quality of self-service system will be influenced whether when the blurring of photos of camera shooting, is led to
Whether the picture for crossing camera shooting clearly determines whether camera can be used, and is a kind of realistic criterion.
The step of various methods divide above, be intended merely to describe it is clear, when realization can be merged into a step or
Certain steps are split, multiple steps are decomposed into, as long as including identical logical relation, all in the protection scope of this patent
It is interior；To adding inessential modification in algorithm or in process or introducing inessential design, but its algorithm is not changed
Core design with process is all in the protection scope of the patent.
Four embodiment of the invention is related to a kind of device for detecting camera status, as shown in Figure 4, comprising:
Module 401 is obtained, for obtaining the picture for the N seed type that camera takes, wherein N is the natural number greater than 2；
Training module 402, for carrying out convolutional neural networks model training for the picture of N seed type as training data；Prediction module
403, the type for the picture using trained model prediction camera captured in real-time；Determination module 404, for according to prediction
Whether result judgement camera can be used.
In one example, above-mentioned training module 402 is specifically used for selecting multiple convolutional neural networks；By N seed type
Picture carries out model training to each convolutional neural networks in multiple convolutional neural networks as training data；Above-mentioned prediction module
403, specifically for the type of the picture using trained each convolutional neural networks model prediction camera captured in real-time；According to each
The prediction result of convolutional neural networks model is voted, and final prediction result is obtained.
In one example, above-mentioned multiple convolutional neural networks include at least depth residual error network；It is above-mentioned according to each convolution
The prediction result of neural network is voted, and final prediction result is obtained, comprising: if voting results are flat ticket, with depth
The prediction result of residual error network is as final prediction result.
In one example, above-mentioned determination module 404 is specifically used for judging whether prediction result characterizes camera captured in real-time
Picture it is clear, if so, determining that camera is available, otherwise, it is determined that camera is unavailable.
In one example, the type of the above-mentioned picture using trained model prediction camera captured in real-time is specific to wrap
It includes: utilizing the type for the L picture that trained model prediction camera is shot recently, wherein L is the natural number greater than 1；It is above-mentioned
Judge prediction result whether characterize camera captured in real-time picture it is clear, if so, determining that camera is available, otherwise, it is determined that camera
It is unavailable, specifically: it is clear to judge whether prediction result characterizes L picture, if so, determine that camera is available, otherwise, it is determined that
Camera is unavailable.
In one example, after above-mentioned judgement camera is unavailable, then the down state of camera is reported.
In one example, above-mentioned N is 6, above-mentioned N seed type, specifically: empty clear type, empty vague category identifier, most
Clear type, most vague category identifier expose clear type, expose vague category identifier.
It is not difficult to find that present embodiment be with first embodiment to the corresponding system embodiment of third embodiment,
Present embodiment can work in coordination implementation with first embodiment to third embodiment.First embodiment is to third embodiment party
The relevant technical details mentioned in formula are still effective in the present embodiment, and in order to reduce repetition, which is not described herein again.Accordingly
Ground, the relevant technical details mentioned in present embodiment are also applicable in first embodiment into third embodiment.
It is noted that each module involved in present embodiment is logic module, and in practical applications, one
A logic unit can be a physical unit, be also possible to a part of a physical unit, can also be with multiple physics lists
The combination of member is realized.In addition, in order to protrude innovative part of the invention, it will not be with solution institute of the present invention in present embodiment
The technical issues of proposition, the less close unit of relationship introduced, but this does not indicate that there is no other single in present embodiment
Member.
Fifth embodiment of the invention is related to a kind of electronic equipment, as shown in Figure 5, comprising:
At least one processor 501；And
With the memory 502 of at least one processor 501 communication connection；Wherein,
Memory 502 is stored with the instruction that can be executed by least one processor 501, instructs by least one processor
501 execute, so that the method that at least one processor 501 is able to carry out the detection camera status such as preceding claim.
Wherein, memory 502 is connected with processor 501 using bus mode, and bus may include any number of interconnection
Bus and bridge, bus is by one or more processors 501 together with the various circuit connections of memory 502.Bus may be used also
With by such as peripheral equipment, voltage-stablizer, together with various other circuit connections of management circuit or the like, these are all
It is known in the art, therefore, it will not be further described herein.Bus interface provides between bus and transceiver
Interface.Transceiver can be an element, be also possible to multiple element, such as multiple receivers and transmitter, provide for
The unit communicated on transmission medium with various other devices.The data handled through processor 501 pass through antenna on the radio medium
It is transmitted, further, antenna also receives data and transfers data to processor 501.
Processor 501 is responsible for management bus and common processing, can also provide various functions, including timing, periphery connects
Mouthful, voltage adjusting, power management and other control functions.And memory 502 can be used for storage processor 501 and execute
Used data when operation.
Sixth embodiment of the invention is related to a kind of computer readable storage medium, is stored with computer program.Computer
Above method embodiment is realized when program is executed by processor.
That is, it will be understood by those skilled in the art that implement the method for the above embodiments be can be with
Relevant hardware is instructed to complete by program, which is stored in a storage medium, including some instructions are to make
It obtains an equipment (can be single-chip microcontroller, chip etc.) or processor (processor) executes each embodiment method of the application
All or part of the steps.And storage medium above-mentioned includes: USB flash disk, mobile hard disk, read-only memory (ROM, Read-
OnlyMemory), random access memory (RAM, RandomAccessMemory), magnetic or disk etc. are various can store
The medium of program code.
It will be understood by those skilled in the art that the respective embodiments described above are to realize specific embodiments of the present invention,
And in practical applications, can to it, various changes can be made in the form and details, without departing from the spirit and scope of the present invention.
Claims (10)
1. a kind of method for detecting camera status characterized by comprising
Obtain the picture for the N seed type that camera takes, wherein N is the natural number greater than 2；
Convolutional neural networks model training is carried out using the picture of the N seed type as training data；
Utilize the type of the picture of trained model prediction camera captured in real-time；
Determine whether camera can be used according to prediction result.
2. the method for detection camera status according to claim 1, which is characterized in that the figure by the N seed type
Piece carries out convolutional neural networks model training as training data, specifically includes:
Select multiple convolutional neural networks；
Using the picture of the N seed type as training data to each convolutional neural networks in the multiple convolutional neural networks into
Row model training；
The type of the picture using trained model prediction camera captured in real-time, specifically includes:
Utilize the type of the picture of camera captured in real-time described in trained each convolutional neural networks model prediction；
It is voted according to the prediction result of each convolutional neural networks model, obtains final prediction result.
3. the method for detection camera status according to claim 2, which is characterized in that
The multiple convolutional neural networks include at least depth residual error network；
The prediction result according to each convolutional neural networks is voted, and final prediction result is obtained, comprising:
If the voting results are flat ticket, using the prediction result of depth residual error network as final prediction result.
4. the method for detection camera status according to claim 1, which is characterized in that described to determine phase according to prediction result
Whether machine can be used specifically:
Judge the prediction result whether characterize the camera captured in real-time picture it is clear, if so, determining that the camera can
With otherwise, it is determined that the camera is unavailable.
5. the method for detection camera status according to claim 4, which is characterized in that described pre- using trained model
The type for surveying the picture of camera captured in real-time, specifically includes:
Utilize the type for the L picture that camera described in trained model prediction is shot recently, wherein L is the nature greater than 1
Number；
It is described judge the prediction result whether characterize the camera captured in real-time picture it is clear, if so, determining the phase
Machine is available, otherwise, it is determined that the camera is unavailable, specifically:
It is clear to judge whether prediction result characterizes the L picture, if so, determining that the camera is available, otherwise, it is determined that institute
It is unavailable to state camera.
6. the method for detection camera status according to claim 4, which is characterized in that determine that the camera can not described
With rear, the down state of the camera is reported.
7. the method for detection camera status according to claim 1, which is characterized in that the N is 6, the N seed type,
Specifically:
Empty clear type, empty vague category identifier, clearest type, most vague category identifier expose clear type, expose fuzzy class
Type.
8. a kind of device for detecting camera status characterized by comprising
Module is obtained, for obtaining the picture for the N seed type that camera takes, wherein N is the natural number greater than 2；
Training module, for carrying out convolutional neural networks model training for the picture of the N seed type as training data；
Prediction module, the type for the picture using trained model prediction camera captured in real-time；
Determination module, for determining whether camera can be used according to prediction result.
9. a kind of electronic equipment characterized by comprising
At least one processor；And
The memory being connect at least one described processor communication；Wherein,
The memory is stored with the instruction that can be executed by least one described processor, and described instruction is by described at least one
It manages device to execute, so that at least one described processor is able to carry out the detection camera status as described in any in claim 1 to 7
Method.
10. a kind of computer readable storage medium, is stored with computer program, which is characterized in that the computer program is located
Reason device realizes the method that camera status is detected described in any one of claims 1 to 7 when executing.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811377288.9A CN109167998A (en) | 2018-11-19 | 2018-11-19 | Detect method and device, the electronic equipment, storage medium of camera status |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811377288.9A CN109167998A (en) | 2018-11-19 | 2018-11-19 | Detect method and device, the electronic equipment, storage medium of camera status |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109167998A true CN109167998A (en) | 2019-01-08 |
Family
ID=64875027
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811377288.9A Pending CN109167998A (en) | 2018-11-19 | 2018-11-19 | Detect method and device, the electronic equipment, storage medium of camera status |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109167998A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111126206A (en) * | 2019-12-12 | 2020-05-08 | 创新奇智(成都)科技有限公司 | Smelting state detection system and method based on deep learning |
| CN111246203A (en) * | 2020-01-21 | 2020-06-05 | 上海悦易网络信息技术有限公司 | Camera blur detection method and device |
| CN111935480A (en) * | 2020-08-03 | 2020-11-13 | 深圳回收宝科技有限公司 | Detection method for image acquisition device and related device |
| CN113055570A (en) * | 2021-03-09 | 2021-06-29 | 广东便捷神科技股份有限公司 | Visual identification method for improving commodity information |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102750824A (en) * | 2012-06-19 | 2012-10-24 | 银江股份有限公司 | Urban road traffic condition detection method based on voting of network sorter |
| CN106600581A (en) * | 2016-12-02 | 2017-04-26 | 北京航空航天大学 | Train operation fault automatic detection system and method based on binocular stereoscopic vision |
| CN107705334A (en) * | 2017-08-25 | 2018-02-16 | 北京图森未来科技有限公司 | A kind of video camera method for detecting abnormality and device |
| CN108229240A (en) * | 2016-12-09 | 2018-06-29 | 杭州海康威视数字技术股份有限公司 | A kind of method and device of determining picture quality |
-
2018
- 2018-11-19 CN CN201811377288.9A patent/CN109167998A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102750824A (en) * | 2012-06-19 | 2012-10-24 | 银江股份有限公司 | Urban road traffic condition detection method based on voting of network sorter |
| CN106600581A (en) * | 2016-12-02 | 2017-04-26 | 北京航空航天大学 | Train operation fault automatic detection system and method based on binocular stereoscopic vision |
| CN108229240A (en) * | 2016-12-09 | 2018-06-29 | 杭州海康威视数字技术股份有限公司 | A kind of method and device of determining picture quality |
| CN107705334A (en) * | 2017-08-25 | 2018-02-16 | 北京图森未来科技有限公司 | A kind of video camera method for detecting abnormality and device |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111126206A (en) * | 2019-12-12 | 2020-05-08 | 创新奇智(成都)科技有限公司 | Smelting state detection system and method based on deep learning |
| CN111246203A (en) * | 2020-01-21 | 2020-06-05 | 上海悦易网络信息技术有限公司 | Camera blur detection method and device |
| WO2021147383A1 (en) * | 2020-01-21 | 2021-07-29 | 上海万物新生环保科技集团有限公司 | Camera blur detection method and apparatus |
| CN111935480A (en) * | 2020-08-03 | 2020-11-13 | 深圳回收宝科技有限公司 | Detection method for image acquisition device and related device |
| WO2022027816A1 (en) * | 2020-08-03 | 2022-02-10 | 深圳回收宝科技有限公司 | Detection method for image acquisition apparatus, and related apparatus |
| CN111935480B (en) * | 2020-08-03 | 2022-02-15 | 深圳回收宝科技有限公司 | Detection method for image acquisition device and related device |
| CN113055570A (en) * | 2021-03-09 | 2021-06-29 | 广东便捷神科技股份有限公司 | Visual identification method for improving commodity information |
| CN113055570B (en) * | 2021-03-09 | 2022-10-18 | 广东便捷神科技股份有限公司 | Visual identification method for improving commodity information |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109167998A (en) | Detect method and device, the electronic equipment, storage medium of camera status | |
| CN105938559B (en) | Use the Digital Image Processing of convolutional neural networks | |
| CN108229307A (en) | For the method, apparatus and equipment of object detection | |
| CN109345260A (en) | A kind of fraud detection model training method and device and fraud detection method and device | |
| CN104063686B (en) | Crop leaf diseases image interactive diagnostic system and method | |
| CN109201514A (en) | Waste sorting recycle method, garbage classification device and classified-refuse recovery system | |
| CN113159300B (en) | Image detection neural network model, training method thereof and image detection method | |
| CN110473083A (en) | Tree-shaped adventure account recognition methods, device, server and storage medium | |
| CN107690660A (en) | Image-recognizing method and device | |
| CN111382808A (en) | Vehicle detection processing method and device | |
| CN108427927A (en) | Target recognition methods and device, electronic equipment, program and storage medium again | |
| CN109815770A (en) | Two-dimentional code detection method, apparatus and system | |
| CN108460362A (en) | A kind of system and method for detection human body | |
| CN106355367A (en) | Warehouse monitoring management device | |
| CN108256404A (en) | Pedestrian detection method and device | |
| CN110059646B (en) | Method for training action planning model and target searching method | |
| CN110069959A (en) | A kind of method for detecting human face, device and user equipment | |
| WO2021027157A1 (en) | Vehicle insurance claim settlement identification method and apparatus based on picture identification, and computer device and storage medium | |
| CN109919252A (en) | The method for generating classifier using a small number of mark images | |
| CN113743417B (en) | Semantic segmentation method and semantic segmentation device | |
| CN110276257A (en) | Face identification method, device, system, server and readable storage medium storing program for executing | |
| CN110084113A (en) | Biopsy method, device, system, server and readable storage medium storing program for executing | |
| CN112016617B (en) | Fine granularity classification method, apparatus and computer readable storage medium | |
| CN116385430A (en) | Machine vision flaw detection method, device, medium and equipment | |
| CN111126264A (en) | Image processing method, device, equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190108 |